Multicyclic sediment transfer along and across convergent plate boundaries (Barbados, Lesser Antilles) Mara Limonta, * Eduardo Garzanti, * Alberto Resentini, * Sergio And o, * Maria Boni and Thilo Bechst adt , § *Laboratory for Provenance Studies, Department of Earth and Environmental Sciences, Universit a di Milano- Bicocca, Milano, Italy Dipartimento di Scienze della Terra, dell’Ambiente e delle Risorse, Universit a Federico II di Napoli, Napoli, Italy GeoResources STC, Heidelberg, Germany §Institute of Geological Sciences, Jagiellonian University, Krakow, Poland ABSTRACT The main source of siliciclastic sediment in the Barbados accretionary prism is off-scraped quartzose to feldspatho-litho-quartzose metasedimentaclastic turbidites, ultimately supplied from South America chiefly via the Orinoco fluvio-deltaic system. Modern sand on Barbados island is either quartzose with depleted heavy-mineral suites recycled from Cenozoic turbidites and including epi- dote, zircon, tourmaline, andalusite, garnet, staurolite and chloritoid, or calcareous and derived from Pleistocene coral reefs. The ubiquitous occurrence of clinopyroxene and hypersthene, associated with green-brown kaersutitic hornblende in the north or olivine in the south, points to reworking of ash-fall tephra erupted from andesitic (St Lucia) and basaltic (St Vincent) volcanic centres in the Lesser Antilles arc. Modern sediments on Barbados island and those shed by larger accretionary prisms such as the Indo-Burman Ranges and Andaman-Nicobar Ridge define the distinctive miner- alogical signature of Subduction Complex Provenance, which is invariably composite. Detritus recy- cled from accreted turbidites and oceanic mudrocks is mixed in various proportions with detritus from the adjacent volcanic arc or carbonate reefs widely developed at tropical latitudes. Ophiolitic detritus, locally prominent on the Andaman Islands, is absent on Barbados, where the prism formed above a westward subduction zone with a shallow decollement plane. The four-dimensional com- plexities inherent with multicyclic sediment dispersal along and across convergent plate boundaries require quantitative provenance analysis as a basic tool in paleogeographic reconstructions. Such analysis provides the link between faraway factories of detritus and depositional sinks, as well as clues on subduction geometry and the nature of associated growing orogenic belts, and even information on climate, atmospheric circulation and weathering intensity in source regions. The beasts that talk, The streams that stand, The stones that walk, The singing sand... Josephine Tey, Singing sands INTRODUCTION Sediments sourced in large orogenic belts generated by oceanic or continental subduction are conveyed long- distance by major river systems across foreland basins, and eventually supplied to continental margins at spe- cific deltaic or estuarine entry points (Potter, 1978; Dickinson, 1988; Hinderer, 2012). Sediment dispersal continues via turbidity currents for hundreds to thou- sands of kilometers beyond the river mouth, and huge masses of sediment are thus transferred from the conti- nent to the deep ocean (Ingersoll et al., 2003). This typ- ically occurs along the trend of major Himalayan-type continentcontinent collision zones, where huge turbi- ditic successions accumulate on remnant-ocean floors destined to be subsequently subducted, while the clastic cover is detached and progressively accreted at the front of a growing fold-thrust belt (Fig. 1a; Morley et al., 2011). Geologically and geometrically distinct is the case of the Caribbean accretionary prism (Fig. 1b). Here, detritus generated in the Andean Cordillera and carried along the retro-belt basin by the Orinoco River finally reaches the Atlantic Ocean, and is deposited by turbidity Correspondence: Eduardo Garzanti, Laboratory for Provenance Studies, Department of Earth and Environmental Sciences, Universit a di Milano-Bicocca, 20126 Milano, Italy. E-mail: eduardo.garzanti@unimib.it © 2014 The Authors Basin Research © 2014 John Wiley & Sons Ltd, European Association of Geoscientists & Engineers and International Association of Sedimentologists 696 Basin Research (2015) 27, 696–713, doi: 10.1111/bre.12095 EAGE